72 BUREAU OF AMERICAN ETHNOLOGY [Boll. 192 



ordinary canvas wall tents that were used at the same camps. There 

 seems to have been a smoke hole in the roof, sometimes equipped with 

 a windscreen. The sheets of bark were laid over a light framework 

 of poles, and might run either vertically or horizontally for the walls. 

 The bark was held down by poles, oars, or sealskins stretched in their 

 drying frames. The door at the front was a blanket or piece of canvas. 



At fishing camps the Yakutat people had smokehouses which not 

 only served to cure the catch but to house the owners. There were 

 separate sleeping rooms in some, and such houses must have been 

 very similar to the smaller houses at the winter villages. Modern 

 smokehouses have either a single ridgepole or twin ridgepoles, although 

 the people live in separate shacks at the fish camps. Whereas the 

 older smokehouses have movable windscreens, modern ones usually 

 have fixed chimneylike types of wood built around the smoke holes, 

 probably copied from the Russians. In the old days there was a 

 false ceiling inside over the fire to spread the smoke and catch the soot, 

 and above this were several tiers of drying racks. Most modern 

 smokehouses lack the false ceiling. 



An informant described caches in the Situk-Lost River area of 

 about 1850 to 1885. These were big holes in the groimd, dug deep so 

 that the food inside would not freeze, and were shored up or lined 

 with logs so that sand would not cave in. Over the pit was a tentlike 

 roof of planks, with a small entrance at one end. A ladder was used 

 to reach the bottom of the cache. Such storehouses were built behind 

 the dwellings. 



A second informant said that caches in the Dry Bay area, dating 

 from about 1900, were only 1 foot deep. Logs were put around the 

 sides, like a log cabin. The roof of bark was sometimes arched as 

 on a Quonset hut, 5 feet high. Walls and roof were covered with 

 mud to protect the contents of the cache from the cold. 



Both informants agreed that each housewife in the multifamily 

 dwelling had her own space in the cache where she kept her bundles 

 of dried fish and boxes of meat and berries preserved in seal oil. These 

 descriptions would explain many of the pits at Knight Island and at 

 other sites. 



We were also told that when eulachon were caught, the fish were 

 put into pits lined with leaves to keep out the sand, and were allowed 

 to rot a little, otherwise the oil, when rendered in a dugout with hot 

 rocks, would not keep. This explanation might serve for some pits 

 on Lost River and Situk River, where there were eulachon runs, but 

 probably not for any on Knight Island, which was perhaps too far 

 from any fish stream. Salmon heads and fish eggs were, however, 

 often buried in the ground to ferment. 



